ANALYSES OF TECHNOLOGICAL AND POLICY OPTIONS FOR ADAPTATION TO CONSEQUENCES OF CLIMATE CHANGE

Transcription

1 ANALYSES OF TECHNOLOGICAL AND POLICY OPTIONS FOR ADAPTATION TO CONSEQUENCES OF CLIMATE CHANGE OVERVIEW OF AGRO-ECOLOGICAL ZONES ADAPTATION: THE CASE OF CROPS AND LIVESTOCK By Hubert E. Meena Maynard S. Lugenja O. A. Ntikha and Mariana Hermes March 28

2 TABLE OF CONTENTS 1. INTRODUCTION AND BACKGROUND INFORMATION INTRODUCTION BACKGROUND INFORMATION STUDY AREAS METHODOLOGY LITERATURE REVIEW TREND ANALYSIS CONSULTATIONS CLIMATOLOGICAL AND ENVIRONMENTAL FACTORS CROP PRODUCTION Rainfall analysis Temperature analysis Area wise Agro Ecological Zone (AEZ) trend analyses for rainfall and maximum temperatures and implication to crop production LIVESTOCK PRODUCTION Tanga Area (AEZ 1) Trend Analyses Dodoma Area (AEZ II) Trend Analyses Shinyanga Area (AEZ III) Trend Analyses Mbeya Area (AEZ IV & V) Trend Analyses Kilimanjaro Area (AEZ VI) Trend Analyses Rufiji Area (AEZ VII) Trend Analyses CLIMATE CHANGE IMPACTS CROP SECTOR Evidence of Climate Change Evidence of AEZ Shifts Way Forward LIVESTOCK SECTOR Trends in Cattle Population and Average Rainfall Trends in Cattle Population and Mean Maximum Temperatures Trends in Cattle Population and Mean Minimum Temperatures Ranking Agro Ecological Zone Areas WAY FORWARD ADAPTATION STRATEGIES ADAPTATION STRATEGIES FOR CROP SECTOR ADAPTATION STRATEGIES FOR THE LIVESTOCK SECTOR Livestock farmers Planned adaptation Increasing awareness through education CONCLUSIONS AND RECOMMENDATIONS CONCLUSIONS Crop Sector Livestock Sector RECOMMENDATIONS ii

3 6.2.1 Crop Sector Livestock Sector REFERENCES... 3 iii

4 1. INTRODUCTION AND BACKGROUND INFORMATION 1.1 Introduction In most African countries, farming depends entirely on the quality of the rainy season-a situation that makes Africa particularly vulnerable to climate change. About million additional people may be at risk of hunger with a temperature increase of 3 C, with more than half of these people concentrated in Africa and Western Asia (IPCC, 21). Dwindling food security provides yet another manifestation of negative effects of climate change in Tanzania. Following persistent drought around steadily rising temperatures, crop failure has been a common phenomenon particularly among seasonal food crops, the situation which, with appropriate modeling predicts even worse food security situation in most semi-arid and arid Tanzania. Livestock is at risk with animals dying alongside deteriorating pasture condition and drying water sources. In recent Years Lake Victoria experienced falling water levels down by almost 2 m. Underground water levels have been falling increasing cost of acquiring it. Rains are running away and cycles are detrimentally changing. While nearly three quarters of Africa relies on agriculture, the most vulnerable way of life as far as climate change is concerned; in Tanzania 37 districts (about one third of the country) are at risk. The famous snow cap sourcing most of rivers and streams in Kilimanjaro region and North Eastern Highlands has been gradually disappearing and surrounding agricultural activities getting negatively impacted increasing concern by Government, local communities and world of tourism all alike as the natural beauty of the great Mt. Kilimanjaro fades away. 1.2 Background Information Tanzania is one of the countries in Greater Horn of Africa; with several other countries. These countries are characterized by chronic food insecurity caused mainly by land degradation. All the countries in the region have faced famines or major crises in the past four years with more localized crises in Tanzania on The situation has been compounded by social and political instability; together with persistent poverty ( problems are concentrated in critical resource areas (transboundary regions). In this region the average productivity levels of agriculture (per unit of land) has declined. Although several countries in the region have increased the land area and earnings in export-oriented agriculture in recent years, the strategy did not help the countries and communities to achieve food security. In Tanzania features of the food emergency have been due to drought, poor food distribution and severe food insecurity resulting in dislocation and economic decline ( 1

5 Figure 1: shows that total food supply satisfied requirements for 26/7 in 16 regions Source: MAFSC, 26. Tanzania Government faced with recurrent food problems enacted a Food Security Act in The Act main objective was to maintain food security in the country through coping mechanisms including sales of livestock in the predominant economic activity. Other forms of strategies are also adopted as and when necessary according to Crop Monitoring and Early Warning Unit (CMEWU, 21) of the food security Department in the Ministry of Agriculture. But when the food situations are beyond control, food aid is adopted where Government uses its SGR to cushion food insecurity on the short-term basis. Under grave situations donor community is called upon to give a hand in dealing with the situation (CMEWU, 21). Data shows that in 6 out of the last 1 years SSR has exceeded 1% implying a selfsufficiency level that varied among the different years between 2% in 22/3 and 17% in 1996/97. On the other hand 4 years have had SSR below 1% by between 6% in 21/2 and 12% in 1997/98. The extent of current year s self sufficiency level is 1% above 1% surpassed only by that of 1996/97 (MAFSC, 26). 2

6 In view of this situation the government has launched several reforms over time to revamp agriculture in the country. These reforms include: International Monetary Fund (IMF) structural adjustment program (SAP), Tanzania government adjustment program and National Development Vision 225. Although different coping mechanisms including livestock sales in the predominant economic activity were recommended after the establishment of the food security Act in 1991, to date revival of the agricultural sector remains unsuccessful. As a result, there has been a decline in foreign exchange earnings, general decline in the exports and increased food insecurity. In a special programme for food security (SPFS) launched in 1996 by FAO (2) working with Farmers Groups in demonstration plots, showed that yields of rain fed rice increased from 12 to 21 bags/acre (2.7 to 4.7 metric tones/ha) and even in low rainfall years, average yields in the demonstration plots were 2 to 3 percent higher than in the control plots. To complement these improvements in crop production, the SPFS has also been successful in introducing and improving livestock raising (FAO, 21). In this paper we are investigating the effect of climate change on main staple seasonal food crops and propose adaptive measures that could be adopted to reduce vulnerability in Agriculture and Food Security in the affected areas. 1.3 Study Areas In general based on weather, soil fertility and food crop production there are 7 main agro ecologies. This study explores these agro ecologies zones (AEZ) in general and each AEZ in particular to expose evidence of climate change and the effect to food production and food security in the short and long term perspectives. Based on main AEZs of Tanzania a generalistic perspective is made in respect of rainfall and temperatures (maximum) and implication to maize, normally a weather vulnerable crop and sorghum, normally a relatively weather stable crop with a reasonable contrast as far as rainfall and temperatures are concerned. This broad view of the existence of climate change effect is explored to reveal a variation over time and over season. Interest is to finally lay foundation for recommending best agricultural practices to adapt to looming climate change and boost production to ascertain food security. Representative areas in each of the seven agro-ecological zones have been used in this study. The areas are shown in Table 1. Table 1: Study Areas Zone Sub-zone Area Characteristics of the zone Coastal-I North Tanga Infertile sands on gently rolling uplands Under 3 m Above sea level; Bi-modal rainfal mm Growing season-december and March -June Arid Lands-II South Dodoma Rolling plains of low soil fertility susceptible to water erosion 5-15 m; unimodal and unreliable-4-6 mm 3

7 Growing season-march to May S/A lands-iii Central Shinyanga Undulating plains with rocky hills and low scarps. Well drained soils with low fertility etc;1-12 m Growing season-december to March Plateau-IV Western Mbeya Wide sandy plains and rift valley scarps; unimodal 8-1 mm; Growing season is November - April S and W Highlands-V Southern Mbeya Undulating plains to dissected hills and mountains. Moderately to fertile clay and volcanic soils Northern Highlands-VI Northern Kilimanjaro Granite Mts in Kilimanjaro; 1-2m; bi-modal and very reliable 1-2 mm; growing season is October - December and March - June Alluvial Plains-VII Coast Rufiji Wide mangrove swamp delta, alluvial soils, sandy upstream, loamy down stream in flood plain 2. Methodology Based on available data and information from secondary sources AEZ approach is considered most effective way of detecting effect of climate change in Agriculture and a suitable starting point towards recommending means and ways for adaptation. Ministry of Agriculture which, is in the forefront in managing agriculture and food security within natural environmental (agro-ecological) settings has been most resourceful in providing appropriate food crop production data and information for this purpose and TMA will be the best provider of climate data. All aspects of analysis will assume AEZ in general and individual AEZ in particular as we focus towards detecting evidence of climate change in individual agro-ecologies. Representative study areas have been carefully selected mainly taking cognizance of data availability and reliability. 2.1 Literature review Various literatures with regard to food, cash crop and livestock production problems in Tanzania have been reviewed. The purpose was to obtain evidence of the existence of climate change impacts per se in Tanzania by using crop production and livestock data for the past 2-years. Secondary data was obtained from national reports, scientific materials, publications and Internet and were summarized accordingly 2.2 Trend analysis An analysis of rainfall, maximum temperatures, crop production (tonnage) and livestock population has been attempted for maize, sorghum and livestock numbers. In this study n=2 years observations have been used in respect of rainfall, maximum temperature, livestock population, maize and sorghum production. For the purpose of this secondary data based study trend analysis is adequate to explore the effect of climate variability overtime and manifestation of climate change in Tanzanian agriculture and livestock sectors. 2.3 Consultations 4

8 Consultations were also done with relevant officers in the Ministry of Agriculture for correct crop production data and filling gaps where necessary. The Statistical Unit in the Ministry of Agriculture was quite resourceful in this endeavor. 3. CLIMATOLOGICAL AND ENVIRONMENTAL FACTORS 3.1 Crop Production Rainfall analysis With results in the foregoing section, analysis of rainfalls performance over the past three decades , and in the Tanzania AEZ is shown in Tables 2 and 3. Table 2: The AEZ of Tanzania Environmental Factors (EF) and Food Crop Production AEZ s/n Name Representative area EF-1 EF-2 Crop 1 Crop 2 AEZ 1 Coastal Tanga Rainfall Maximum Temperatures AEZ 2 Arid land Dodoma Rainfall Maximum Temperatures AEZ 3 Semi-Arid lands Shinyanga Rainfall Maximum Temperatures Maize Maize Maize Sorghum Sorghum Sorghum AEZ 4 Plateau Mbeya Rainfall Maximum Temperatures Maize Sorghum Table 3: analysis of rainfalls performance over the past three decades , and in the Tanzania Agro Ecological Zones AEZs Study area Estimated model AEZ-I Coastal y = -7.52x x AEZ-II Arid land-dodoma: y = x x AEZ-III Semi-Arid lands- Shinyanga y = x x AEZ-IV Plateau-Mbeya y = x x AEZ-V South Western Highlands- Mbeya y = x x AEZ-VI Northern Highlands -Kilimanjaro y = 1.433x x AEZ-VII Alluvial-Morogoro y = x x In general, except for Morogoro and Moshi in Kilimanjaro region all zones reflect negative sloping trends implying falling rainfall availability overtime (Figure 3). In view of this there is need to adapt to these rainfall shortages probably by adopting water conserving technologies. Water harvesting initiatives should not be taken for granted. 5

9 An analysis of long term means of monthly availability of rains across AEZ show that on average total rains for a typical crop season is normally highest in Tanga followed by Mbeya, Kilimanjaro, Shinyanga Morogoro and Dodoma. Rainfall intensity depends on bimodality tendency, which may be best described by proportionate contribution of each season to crop production. Thus Tanga ranks top in rainfall distribution over all 12 months followed by Mbeya, Kilimanjaro and Morogoro. Dodoma is the weakest distributed region with 4 months (June, July, August and September) having no rains at all followed by Shinyanga where 3 months (June, July and August) have virtually no rains (Figure 3). Figure 3: Rainfalls performance over the past three decades , and in the Tanzania AEZ y = -7.52x x R 2 = 1 y = x x R 2 = 1 Dodoma Met Kilimanjaro Airport Mbeya Met Morogoro Met Moshi Met Same Met R ainfall (m m) y = 1.433x x R 2 = 1 y = x x R 2 = 1 y = x x R 2 = 1 Tanga met Shinyanga Met Poly. (Tanga met) Poly. (Mbeya Met) 4. y = 2.696x x R 2 = 1 y = x x R 2 = 1 y = x x R 2 = 1 Poly. (Moshi Met) Poly. (Shinyanga Met) Poly. (Morogoro Met) 2. Poly. (Dodoma Met) Poly. (Same Met) Decadal years Poly. (Kilimanjaro Airport) An analysis of monthly rainfall maxima across AEZs over the 2 years of this study (1984/85-23/4) shows that only Dodoma did not receive any rains at all in September and July. All AEZs have rains in March and April The range between highest and lowest rainfall values shows that highest range was found in Tanga followed by Mbeya, Kilimanjaro, Shinyanga, Dodoma and Morogoro. The implication to maize (vulnerable crop) and sorghum (relatively tolerant crop) has been worked out and presented later in this section. 6

10 3.1.2 Temperature analysis An analysis of maximum temperature performance over the past three decades , and in the Agro-ecological zones shows that in general, except for Same in Kilimanjaro region, all areas indicate an increasing temperature of at least.2% per decade implying an increase of at least 2% in 1 years time! Analysis of monthly long term means of maximum temperatures per AEZ across crop season shows that the hottest months with temperatures above 3 o C were October to March with peak values (33 o C) in February in Kilimanjaro. The situation was coldest (as low as 21.6 o C) in AEZ IV and V. Maximum temperatures over the 2 years of the study could reach 35.7 o C in AEZ IV. The result shows that the range between minimum and maximum temperatures was highest (reached 1.6) in February in AEZ VI Area-wise Agro Ecological Zone (AEZ) trend analyses for rainfall and maximum temperatures and implication to crop production In each region representing an AEZ a zoom in analysis of maximum temperatures and rainfalls over the period of study (1984/85-23/4) was undertaken. The analysis focused at monthly maxima over the period, the long term means calculated on the bases of 3 years (1974/75-23/4) on the one hand. On the other hand the analysis focused at maize and later sorghum was made to ascertain the implication of climate factors to crop production at individual AEZ. Accordingly, the following results are organized for Tanga (to represent Coastal zone-i) Mbeya (Plateau-IV, S and W Highlands-V), Shinyanga (S/A lands-iii), Dodoma (Arid Lands-II), Kilimanjaro (Northern Highlands-VI) and Morogoro (Alluvial Plains-VII). In each representative region we explore changes in the two major climatic variables namely rainfall and maximum temperatures with implication to production of two major crops namely maize and sorghum. In AEZ I represented by Tanga (Table 5) both maize and sorghum are positively correlated with climate factors influencing their production. Both have high parameters of association as reflected from the values of R 2, i.e. 66% for maize and 88% for sorghum. In AEZ II represented by Dodoma (Table 5) maize and sorghum are positively correlated with climate factors influencing their production. However they are both not so strongly associated as reflected from the values of R 2 which are 46% for maize and 11% for sorghum. There may be other factors that share the strength of association with rainfall in particular. 7

11 Table 5: A Summary of trend analysis of maize and sorghum production under the influence of rainfall and maximum temperatures over the 2 years of study (1984/85-23/4) AEZ Study area Estimated model R 2 AEZ-I Coastal Rainfall y =.126x x Temperatures y = -7.52x x AEZ-II AEZ-III AEZ-IV AEZ-V AEZ-VI AEZ-VII Maize y =.8485x x Sorghum y =.2226x x Arid land Rainfall y = x x Temperatures y =.1163x x Maize y =.7277x x Sorghum y =.1556x x Semi-Arid lands- Shinyanga Rainfall y = x x Temperatures y =.2557x Maize y = x x Sorghum y = -.946x x Plateau-Mbeya Rainfall y = x x Temperatures y =.1611x x Maize y = x x Sorghum y =.871x x South Western Highlands- Mbeya Rainfall y = x x Temperatures y =.1611x x Maize y = x x Sorghum y =.871x x Northern Highlands -Kilimanjaro Rainfall y = 1.433x x Temperatures y =.33x x Maize y = 1.458x x Sorghum y = -.188x x Alluvial-Morogoro Rainfall y = x x Temperatures y =.57x x Maize y =.7842x x Sorghum y = x x In AEZ III represented by Shinyanga (Table 5) both maize and sorghum are negatively correlated with climate factors influencing their production. However they are both not so strongly associated as reflected from the values of R 2 which are 45% for maize and 23% for sorghum. There may be other factors that share the strength of association with rainfall and temperature. 8

12 In AEZ IV for plateau and V for SW Highlands represented by Mbeya (Table 5) both maize and sorghum are positively correlated with climate factors influencing their production. However they are both not so strongly associated as reflected from the values of R 2 which are 33% for maize and 9% for sorghum. There may be other factors that share the strength of association with rainfall and temperature. In AEZ VI for Northern Highlands represented by Kilimanjaro (Table 5) Maize is positively correlated with climate factors influencing its production while sorghum is negatively correlated. However they are both not so strongly associated as reflected from the values of R 2 which are 23% for maize and 17% for sorghum. There must be other factors that share the strength of association with rainfall and temperature. In AEZ VII for Alluvial plains represented by Morogoro (Table 5) maize is positively correlated with climate factors influencing its production while sorghum is negatively correlated. However they are both not so strongly associated as reflected from the values of R 2 which are 3% for maize and 7% for sorghum. There must be other factors that share the strength of association with rainfall and temperature. In general, it is important to note that the outlying values of 1997/98 during super normal rainfall availability under El-nino will have overestimated the degrees of association. 3.2 Livestock Production Tanga Area (AEZ 1) Trend Analyses Results show that cattle in Tanga Area decreased from 496, heads in 1985 to 34,392 heads in 24. This was about 191,68 (62.95%) fewer cattle than that which existed in A trend shows that this decrease was significant (R 2 =.5446) for the past two decadal years as depicted in Figure 17a. As the cattle numbers decreased progressively, the average rainfall received in the zone insignificantly declined (R 2 =.1355). Except for 1985, 1986, 1987, 1992, 1994,1997,1998,1999 and 22 years which received rainfall above mean decadal rainfall of 19.5 mm, all the other years over two decadal year s period received bellow average rainfall. Similarly as cattle numbers decreased, mean maximum temperatures increased steadily (R 2 =.716) and so were the mean minimum temperatures (R 2 =.546) as shown in Figure 17b. The differences between high and low mean maximum and minimum temperatures over the twenty year period were 1. o C and 1.7 o C as depicted in Figure 17b. 9

13 Trends in Cattle Population and Average Rainfall (mm) in Tanga Area CATTLE NUMBERS 1,, 9, 8, 7, 6, 5, 4, y = -.14x x R 2 =.1355 y = -3161x x R 2 = , 2, 1, YEARS Cattle Population Average Rainfall (mm) Poly. (Average Rainfall (mm)) Poly. (Cattle Population) Figure 17a: Trends in cattle and average rainfall in zone I, Tanga for periods o CENTIGRADE Trends in cattle Population and Mean Maximum Temperature (oc) in Tanga Area CATTLE NUMBERS 1,, 9, 8, 7, 6, 5, 4, 3, 2, 1, y =.2x x R 2 =.716 y = -3161x x R 2 = YEARS o CENTIGRADE Cattle Population Poly. (Mean Max Temp) Mean Max Temp Poly. (Cattle Population) Figure 17b: Trends in cattle and Mean Maximum Temperature in zone I, Tanga for periods Dodoma Area (AEZ II) Trend Analyses In Dodoma Area cattle population decreased from 1,9, heads in 1985 to 798,15 heads in 24. This was about 21,895 (79.1%) fewer cattle than that which existed in A trend shows that this decrease was not significant (R 2 =.4274) for the past two decadal years as depicted Table 18b. As the cattle numbers progressively decreased, the average rainfall received in the zone also significantly declined (R 2 =.6115). Similarly as cattle numbers decreased, mean maximum temperatures increased steadily (r 2 =.652) 1

14 and so were the mean minimum temperatures (r 2 =.115) as shown in the analysis. The differences between high and low mean maximum and minimum temperatures were.8 o C and 1.5 o C as depicted in Figure 18b. Trends in Cattle and Mean Maximum Temperatures (oc) in Dodoma Area CATTLE NUMBERS 1,8, 1,6, 1,4, 1,2, 1,, 8, 6, 4, 2, y = -6224x x R 2 =.4274 y = -.8x x R 2 = YEARS o CENTIGRADE Cattle Population Poly. (Mean Max Temp) Mean Max Temp Poly. (Cattle Population) Figure 18b: Trends in cattle and Mean Maximum Temperature in zone I, Dodoma for period Shinyanga Area (AEZ III) Trend Analyses It is shown in the analysis that in Shinyanga Area, cattle decreased from 1,89,2 heads in 1985 to 3,86,677 heads in 24. This was about 1,916,477 (11.4%) more cattle than that which existed in It shows that this decrease was significant (R 2 =.8646) for the past two decadal years (Table 19a). As the cattle numbers progressively increased, the average rainfall received in the zone also insignificantly declined (R 2 =.98). Similarly as cattle numbers increased, mean maximum temperatures increased steadily (R 2 =.64) and so were the mean minimum temperatures (R 2 =.2581) as shown in Figure 19b. The differences between high and low mean maximum and minimum temperatures were 1.7 o C and 4.5 o C as shown in Figure 19b. 11

15 CATTLE NUMBERS 4,5, 4,, 3,5, 3,, 2,5, 2,, 1,5, 1,, 5, Trends in Cattle Population and Average Rainfall (mm) in Shinyanga Area YEARS y = x x + 2E+6 y =.624x x R 2 =.8646 R 2 =.98 Cattle Population Mean Rainfall Poly. (Mean Rainfall) Poly. (Cattle Population) Figure 19a: Trends in cattle and average rainfall in zone III, Shinyanga for periods o CENTIGRADE CATTLE NUMBERS 4,5, 4,, 3,5, 3,, 2,5, 2,, 1,5, 1,, 5, Trends in Cattle Population and Mean Maximum Temperatures in Shinyanga Area YEARS y = x x + 2E+6 R 2 =.8646 Cattle Population Poly. (Cattle Population) ocentigrade y =.14x x R 2 =.64 Mean Maximum Temperatures Poly. (Mean Maximum Temperatures) Figure 19b: Trends in cattle and Mean Maximum Temperature in zone III, Shinyanga for periods Mbeya Area (AEZ IV & V) Trend Analyses Results show that in Mbeya Area (zone IV and V) cattle decreased from 832,546 heads in 1985 to 819,227 heads in 24. This was about 13,319 (1.41%) fewer cattle than that which existed in A trend shows that this decrease was insignificant (R 2 =.6294) for the past two decadal years as depicted in Figure 2a. As the cattle numbers progressively increased, the average rainfall received in the zone insignificantly declined (R 2 =.143). Similarly as cattle numbers increased, mean maximum temperatures increased steadily (R 2 =.47) and so were the mean minimum temperatures (R 2 =.4864) as shown in Figure 2b. The differences between high and low mean maximum and minimum temperatures were 3.8 o C and 1.5 o C as seen in Figure 2b 12

16 CATTLE NUMBERS 1,2, 1,, 8, 6, 4, 2, Trends in Cattle Population and Average rainfall (mm) in Mbeya Area (zone IV-V) y = -1279x x R 2 =.6294 YEARS y =.2768x x R 2 = Cattle Population Mean Rainfall Poly. (Mean Rainfall) Poly. (Mean Rainfall) Poly. (Cattle Population) Figure 2a: Trends in cattle and average rainfall in zone IV-V-Mbeya for periods ocentigrade CATTLE NUMBERS Trends in Cattle Population and Mean Maximum Temperature in Shinyanga Area 1,2, 1,, 8, 6, 4, 2, o CENTIGRADE y = -1279x x R 2 =.6294 YEARS y = -.11x x R 2 =.47 Cattle Population Poly. (Mean Max Temp) Mean Max Temp Poly. (Cattle Population) Figure 2b: Trends in cattle and Mean Maximum Temperature in zone IV-V-Mbeya for periods Kilimanjaro Area (AEZ VI) Trend Analyses Results show that in Kilimanjaro Area, cattle increased from 414,489 heads in 1985 to 593,273 heads in 24. This was about 178,784 (43.13%) more cattle than that which existed in A trend shows that this decrease was significant (R 2 =.935) for the past two decadal years as depicted in Figure 21a. As the cattle numbers progressively increased, the average rainfall received in the zone insignificantly declined (R2=.476). Similarly as cattle numbers increased, mean maximum temperatures increased steadily (R 2 =.929) and so were the mean minimum temperatures (R 2 =.931) as shown in Figure 21b. The differences between high and low mean maximum and minimum temperatures were 2.1 o C and 1.8 o C as seen in the analysis 13

17 7, Trends in Cattle Population and Average Rainfall (mm) in Kilimanjaro Area 12 CATTLE NUMBERS 6, 5, 4, 3, 2, 1, y = x x R 2 =.935 y = -.678x x R 2 = YEARS ocentigrade Cattle Population Mean Rainfall Poly. (Mean Rainfall) Poly. (Cattle Population) Figure 21a: Trends in cattle and average rainfall in zone VI Kilimanjaro Area for periods Trends in Cattle Population and Mean Maximum Temperatures(oC) in Kilimamnjaro Area CATTLE NUMBERS 7, 6, 5, 4, 3, 2, 1, y = x x R 2 =.935 y =.13x x R 2 = YEARS o CENTIGRADE Cattle Population Poly. (Mean Max Temp) Mean Max Temp Poly. (Cattle Population) Figure 21b: Trends in cattle and Mean Maximum Temperature in zone VI Kilimanjaro Area for periods Rufiji Area (AEZ VII) Trend Analyses Cattle population in Rufiji Area increased from 89 heads in 1985 to 11,64 heads in 24, this was about 124,274 ( 1 %) more cattle than that which existed in A trend shows that this increase was significant (R 2 =.936) for the past two decadal years as depicted in Figure 22a. As the cattle numbers progressively increased, the average rainfall received in the zone insignificantly declined (R 2 =.1892). Similarly as cattle numbers increased, mean maximum temperatures increased steadily (R 2 =.2681) and so were the mean minimum temperatures (R 2 =.5872) as shown in Figure 22b and 22c respectively. The differences between high and low mean maximum and minimum temperatures were 1.8 o C and 1.5 o C as depicted in Figure 22b and 22c respectively. 14

18 CATTLE NUMBERS 16, 14, 12, 1, 8, 6, 4, 2, -2, Trends in Cattle and Averge Rainfall (mm) in zone VII, Rufiji Area y = x x R 2 =.1892 y = x x R 2 = YEARS RAINFALL IN MM Cattle population Average Rainfall (mm) Poly. (Average Rainfall (mm)) Poly. (Cattle population) Figure 22a: Trends in cattle and average rainfall in zone VII Rufiji Area for periods Trends in Cattle Population and Mean Maximum Temperature oc in Rufiji Area CATTLE NUMBERS 16, 14, 12, 1, 8, 6, 4, 2, -2, y =.73x x R 2 =.2681 y = x x R 2 = YEARS o CENTIGRADE Cattle Population Poly. (Mean Maximum Temperature oc) Mean Maximum Temperature oc Poly. (Cattle Population) Figure 22b: Trends in Cattle Population and Mean Maximum Temperature in zone VII Rufiji Area for periods 15

19 Trends in Cattle Population and Mean Minimum Temperatures in Rufuji Area CATTLE NUMBERS 16, 14, 12, 1, 8, 6, 4, 2, -2, y =.11x x R 2 =.5938 y = x x R 2 = YEARS o CNTIGRADE Cattle Population Poly. (Mean Minimum Temperature oc) Mean Minimum Temperature oc Poly. (Cattle Population) Figure 22c: Trends in cattle and Mean Maximum in zone VII Rufiji Area for periods 4. CLIMATE CHANGE IMPACTS 4.1 Crop Sector Evidence of Climate Change Recent evidence of the effect of Climate change on Crop Performance under agroecological influence and observed adaptation efforts. Adaptation to the negative effect of climate change is being exercised here and there in different agro -ecologies. For example in recent MAFSC reports the improvement observed in the production of cassava and potatoes, which are drought tolerant, has been attributed to continued government efforts to vigorously promote such crops (MAFSC, 26). This has been a success story on the one hand. On the other hand, the falling trend in sorghum which is also normally drought tolerant and equally receiving Government promotional efforts needs further explanation. According to field reports, late masika onset is noted as the main reason for decreased production of sorghum due to serious effect of plant pests and diseases such as powdery mildew and head smut as observed during preliminary forecast of food crops in Mwanza region. During the same food crop assessment in Dodoma, excessive heat nearly drove most sorghum to permanent wilting point the situation that affected most of Dodoma rural and Dodoma Urban districts Though maize is more vulnerable in drought prone areas compared to sorghum and millets the mixed status conditions encountered in sorghum and bulrush millets leaves out a strong issue that remains unaddressed. This is probably one of the potential issues that need to be studied under the climate change initiative. Additional evidence of climate change supporting the Shift paradigm is the observed shift of some regions from unimodality to bimodality rainfall regime and vice versa. For example Manyara region which is mainly bordered by unimodal regions of Dodoma and 16

20 Singida as well as the transition region of Morogoro has largely converted from bimodality tendency to unimodality. Also, Morogoro and Kigoma Regions which have been either wholly or partially conceptualized with transition features are increasingly transforming to Unimodal tendencies. Conversely, Mbeya and Shinyanga regions which have traditionally been unimodal are gradually developing bimodality tendencies, for the current status of various regions in Tanzania. At the same time, pasture improvement was observed alongside general recovery of livestock from drought experienced in the earlier part of the 25/6 crop season Box 1 In addition, other factors that affected crop production include insects/pests infestation and unavailability of early maturing seeds. Maximization of existing cropping seasons, introduction of more irrigation schemes, improvement of crop husbandry practices through efficient agricultural extension services, continued expansion of farming areas, raising productivity, improved input supply including high yielding varieties of seeds and improved market information dissemination will continue to demand government intervention if Tanzania is to attain a noble goal of becoming a surplus generator and eventually a net exporter of food. Consistent with these efforts the adaptation to negative effects of climate change which have been largely observed in Dodoma, Singida, Shinyanga, Tabora, Mwanza, Arusha, Kilimanjaro and (Northern parts of) Iringa regions must rank top in the agenda for sustainable crop production. The result of trend analysis for rainfall, maximum temperature, maize and sorghum gives us estimates of the coefficients estimated through a polynomial function. The results ranked in order of time correlation AEZ-VII ranks top followed by AIZ-VI then AEZ-II, AEZ-IV, AEZ-V, AEZ-I and AEZ- III. The positivity in top 2 AEZs shown that there is room for improvement in these AEZs in the shown order of ranks, likewise the negativity observed in Dodoma, Mbeya Tanga and Shinyanga shows the increasing negativity strength in that order. Ranked according to descending a-coefficients AEZ-III ranks top followed by AEZ-IV, AEZ-V, AEZ-II, AEZ-VII and AIZ-VI, the trend values for maize crop are as per Table 8 Table 8: Estimates for maize trend and predicted crop production AEZ-Maize 24/5 (Tones) 214/15 (Tones) Mbeya Kilimanjaro Tanga Morogoro Dodoma Shinyanga

21 Mbeya ranks top followed by Kilimanjaro, Tanga, Morogoro, Dodoma and Shinyanga. Except for AEZ-III, the values are all positive. Consistent with the fitted model, predicted maize production values are as per Table 9 for 24/5 and for 24/5, 1 years after. The forecasts are all positive in the year following years of observation. Likewise, it is all positive ten years after except for AEZ-III. However a critical look into this decadal interval forecasts shows that the forecasts are all increasing with topmost percentage increase in AEZ-VI, followed by AEZ-II, AEZ-I, AEZ-VII, AEZ-IV and AEZ-V. AEZ- III will fall off production and finally in 29/1 at a calculated rate of approximately 21% per annum. Based of trend analysis and the forecasts made thereafter, it is prospective to cultivate maize in the following AEZs (with percentage annual increase in brackets): AEZ-VI (17.6), AEZ-II (14.1), AEZ-I (13.) and AEZ-VII (12.1) and AEZ-IV & AEZ-V (1.9). The policy recommendations go by the same order. As for sorghum the trend coefficients are as per Table 9. Table 9: Estimates for sorghum trend and predicted crop production AEZ-Sorghum 24/5 (Tones) 214/15 (Tones) Percentage Change in Production Dodoma Mbeya Tanga Kilimanjaro Morogoro Shinyanga According to a-coefficients, except for AEZ-2, AEZ-4 and AEZ-5, the rest of AEZs have negative a-coefficients showing that the latter have no definite chance of peaking up as future producers. The b-coefficients are negative for AEZ-2, AEZ-4 and AEZ-5 and positive values for others. The c-coefficients (values of y at -time (origin)) are positive except for AEZ-I and AEZ-VI where negative c-coefficients are recorded. Predicted values are all positive for 24/5 (the first year after years of observation. Ten years later, the forecasts increase in AEZ-I (18.9%), AEZ-IV and AEZ-V (11.2%) and AEZ-II (5.7%). The forecasts decline in AEZ-III (115.4%) and AEZ-VI (16.7%) but stabilize in AEZ-VII. Based on trend analysis and the forecasts made thereafter, it is prospective to cultivate sorghum in AEZ-I, AEZ-IV and AEZ-V and AEZ-II, and the policy recommendations go accordingly Evidence of AEZ Shifts Some of the evidences of Climate change have been pointed out as shifts in agroecological zones (Tanzania NAPA VPO, 25). Change in Crop performance over a period such as a decade may be considered as detective of AEZ shift. Based on estimated coefficients and the forecasts arrived as above (Table 8 & 9) reranking of AEZs has been undertaken on a decadal interval. The results are presented in 18

22 Figure 23 for maize and Figure 24 for sorghum and discussed in line with the data in Table 1 for both maize and sorghum. 1, AEZ-5 8 AEZ-4 AEZ-4 AEZ-3 AEZ-5 AEZ-7 AEZ-5 AEZ-4 6 Production (Tonnes) AEZ-3 AEZ-5 AEZ-7 AEZ-1 AEZ-7 AEZ-1 AEZ-6 AEZ-2 AEZ-1 4 AEZ-2 AEZ-6 AEZ-6 AEZ-6 AEZ-7 2 AEZ-2 AEZ-1 AEZ-3 AEZ / /95 24/5 214/15 Decade 1 Decade 2 Decade 3 Decade Decadal years AEZ-1 AEZ-2 AEZ-3 AEZ-4 AEZ-5 AEZ-6 AEZ-7 Figure 23: Maize forecast over decades Production (Tonnes) 5-5 AEZ-2 AEZ-3 AEZ-2 AEZ-2 AEZ-3 AEZ-2 AEZ-1 AEZ-1 AEZ-5 AEZ-7 AEZ-5 AEZ-4 AEZ-4 AEZ-4 AEZ-5 AEZ-7 AEZ-4 AEZ-7 AEZ-3 AEZ-5 AEZ-1 AEZ-1 AEZ-6 AEZ-6 AEZ-7 AEZ-6 AEZ / /95 24/5 214/ AEZ Decadal years AEZ-1 AEZ-2 AEZ-3 AEZ-4 AEZ-5 AEZ-6 AEZ-7 Figure 24: Sorghum production over decades and the forecast 19

23 Table 1: Predicted values of maize and sorghum production demonstrating shifts in AEZs consistent with climate change effect Maize Decade 1 Decade 2 Decade 3 Decade 4 Maize crop performance demonstrating a shift in AEZs AEZ 1984/ /95 24/5 214/15 AEZ- 1984/85 AEZ- 1994/95 AEZ- 24/5 AEZ- 214/15 AEZ AEZ-4 AEZ-3 AEZ-4 AEZ-4 AEZ AEZ-5 AEZ-4 AEZ-5 AEZ-5 AEZ AEZ-6 AEZ-5 AEZ-7 AEZ-1 AEZ AEZ-3 AEZ-7 AEZ-1 AEZ-7 AEZ AEZ-1 AEZ-1 AEZ-2 AEZ-2 AEZ AEZ-7 AEZ-6 AEZ-6 AEZ-6 AEZ AEZ-2 AEZ-2 AEZ-3 AEZ-3 Sorghu m Decade 1 Decade 2 Decade 3 Decade 4 Sorghum performance demonstrating a shift in AEZs AEZ 1984/ /95 24/5 214/15 AEZ- 1984/85 AEZ- 1994/95 AEZ- 24/5 AEZ- 214/15 AEZ AEZ-2 AEZ-3 AEZ-2 AEZ-2 AEZ AEZ-3 AEZ-2 AEZ-1 AEZ-1 AEZ AEZ-4 AEZ-7 AEZ-4 AEZ-4 AEZ AEZ-5 AEZ-4 AEZ-5 AEZ-5 AEZ AEZ-7 AEZ-5 AEZ-3 AEZ-7 AEZ AEZ-1 AEZ-1 AEZ-7 AEZ-6 AEZ AEZ-6 AEZ-6 AEZ-6 AEZ-3 Initially maize performance (Table 1 and Figure 23) was in the order of AEZ-4, AEZ-5, AEZ-6, AEZ-3, AEZ-1, AEZ-7 and AEZ-2. Ten years later, the order changed to AEZ-3, AEZ-4, AEZ-5, AEZ-7, AEZ-1, AEZ-6 and AEZ-2. Except for AEZ-1 and AEZ-2 the rest of AEZs changed position. In these changes, AEZ-3 and AEZ-7 changed positions upwards while AEZ-4, AEZ-5 and AEZ-6 dropped. In second decade AEZ-4, AEZ-5, AEZ-7, AEZ-1 and AEZ-2 moved up while, AEZ-3 dropped to lowest position leaving AEZ-6 in the same rank. Predicted 1 years later is seen to promote AEZ-1 only leaving AEZ-2, AEZ-3, AEZ-4 and AEZ-5 in the same position and demoting AEZ-7. On the other hand sorghum performed as shown in Table 1 and Figure 24. Initially sorghum production ranked down in the order of AEZ-2, AEZ-3, AEZ-4, AEZ-5, AEZ-7, AEZ-1 and AEZ-6. Ten years later, the order changed to AEZ-3, AEZ-2, AEZ-7, AEZ-4, AEZ-5, AEZ-1 and AEZ-6. Except for AEZ-1 and AEZ-6 the rest of the AEZ changed positions. In these changes, AEZ-3 and AEZ-7 changed positions upwards while, AEZ- 2, AEZ-4 and AEZ-5 dropped. In second decade AEZ-1, AEZ-2, AEZ-4 and AEZ-5 moved up while, AEZ-3 and, AEZ-7 dropped leaving AEZ-6 in the same lowest rank. Predicted third decade is seen to promote AEZ-6 and AEZ-7 leaving AEZ-1, AEZ-2, AEZ-4 and AEZ-5 in the same position and demoting AEZ-3 to the lowest rank. 2

24 Whether these shifts in AEZs provide adequate evidence of climate change is subject to further enquiry which is now being proposed as a way forward Way Forward Based on rank order of climate factors and their influence on crop performance as perceived through modeled coefficients where aggregate ranks have been worked out and average positions ranked as shown in Table 11. Table 11: Aggregate Rank Analysis for Regions representing AEZs AEZ-Regions Rainfall rank Temp rank +maize +sorghum Avg Kilimanjaro Mbeya Dodoma Morogoro Tanga Shinyanga Based on aggregate rank analysis (Table 11), Kilimanjaro area qualifies for further analyses of technological and policy options for adaptation to consequences of climate change followed by Mbeya, Dodoma, Morogoro, Tanga and Shinyanga. Furthermore, socio-economic value attached to Mt. Kilimanjaro adds weight to Kilimanjaro. 4.2 Livestock Sector Trends in Cattle Population and Average Rainfall In Dodoma area 61.15% of the variation in cattle numbers is associated with differences in the amount rainfall received and 38.85% could be accounted for by other factors. (Table 12). Rainfall in Dodoma is usually erratic ( mm) and is confined almost entirely to the period from November to April. This markedly seasonal distribution inevitably produces corresponding fluctuations in herbage availability and nutritive value through the year (Owen, 1964). The area is followed by Rufiji area where 18.91% of the variation in cattle numbers is associated with differences in the amount rainfall received and 81.9% could be accounted for by other factors (Table 12). In Rufiji rainfall ( mm) is not a limiting factor to forage production. The area falls in bi-modal rainfall. The cattle numbers in the area could be controlled by the supply and demand of red meat. Most of the cattle entering the area are usually destined for slaughter in Dar es Salaam. Therefore as human population has increased in city, the cattle population also increased due to high demand for red meat. The least (1.43%) variation in cattle numbers due to rainfall variation is in Mbeya area (Plateau and Southern and Western highlands) as shown in Table 12. Mbeya and Rukwa areas receive a lot of rain (range mm), therefore other factors which accounted for 98.57% of the variation in cattle numbers in these areas could be due to variation in red meat market in neighbouring countries. In other areas (Tanga, Shinyanga and Kilimanjaro) 4.76% % of the 21

25 variation in cattle numbers is associated with differences in the amount rainfall received and 86.45% % could be accounted for by other factors. Table 12: Association of Environmental Factors (Rainfall and Maximum and Minimum Temperatures) on Cattle Population in different Area Zone Sub- Zone Area Av.Rainfall- R 2 Maximum Temp-R 2 Minimum Temp-R 2 Cattle Pop- R 2 Coastal-I North Tanga Arid Lands-II South Dodoma S/A lands-iii Central Shinyanga Plateau-IV Western Mbeya S and W Highlands-V Southern Mbeya Northern Highlands-VI Northern Kilimanjaro Alluvial Plains-VII Coast Rufiji Trends in Cattle Population and Mean Maximum Temperatures In Rufiji area 26.81% of the variation in cattle numbers is associated with differences in the mean maximum temperatures received and 73.19% could be accounted for by other factors (Table 12). Mean maximum temperatures in Rufiji area is 31 o C for and decadal years. This constant ambient temperature depresses, inhibits or slows down the physiology of lower and high animals (Odum, 1971 and Lugenja, 1982). The area is followed by Kilimanjaro area where 9.29% of the variation in cattle numbers is associated with differences in the mean maximum temperatures received and 9.71% could be accounted for by other factors. Mean maximum temperatures in Kilimanjaro area is 29.5 o C for and 29.7 o C for decadal years. The least (.47%) variation in cattle numbers due to mean maximum temperatures variation is in Mbeya area (Plateau and Southern and Western highlands) and 99.53% could be due to other factors. Mean maximum temperatures in Mbeya area is 23.7 o C for and 21.9 o C for decadal years. In other areas (Tanga, Dodoma and Shinyanga) 6.4% % of the variation in cattle numbers is associated with differences in the mean maximum temperatures received and 92.84% % could be accounted for by other factors. Except for Mbeya Area where there is a decrease in mean maximum temperature (change =1.8 o C) in the second decadal year, in other areas (Kilimanjaro, Tanga, Dodoma and Shinyanga) there is an increase in mean maximum temperature (range.1 o C.2 o C) in the second decadal period ( ). These changes in mean maximum temperature though small are quite significant in eliciting climate change in respective areas. According to the Intergovernmental Panel on Climate Change (IPCC, 21) rising temperature and changes in precipitation are undeniably clear with impacts already affecting ecosystems, biodiversity and people and that the negative impacts associated with climate change are also compounded by many factors, including widespread poverty, human diseases, and high population density, which is estimated to double the demand for food, water, and livestock forage within the next 3 years Trends in Cattle Population and Mean Minimum Temperatures In Rufiji area 58.72% of the variation in cattle numbers is associated with differences in the mean minimum temperatures received and 41.28% could be accounted for by other factors (Table 12). Mean minimum temperatures in Rufiji area is 21 o C for and 22

26 22 o C for decadal years. The area is followed by Tanga area where 58.13% of the variation in cattle numbers is associated with differences in the mean minimum temperatures received and 41.87% could be accounted for by other factors. Mean minimum temperatures in Tanga area is 21.9 o C for and 22.6 o C for decadal years. The least (9.31%) variation in cattle numbers due to mean minimum temperatures variation is in Kilimanjaro area and 9.69% could be due to other factors. Mean minimum temperatures in Kilimanjaro area is 17.7 o C for and 18.1 o C for decadal years. In other areas (Mbeya, Shinyanga and Dodoma) 11.5% % of the variation in cattle numbers is associated with differences in the mean minimum temperatures received and 51.36% % could be accounted for by other factors Ranking Agro Ecological Zone Areas Rainfall Mbeya Area ranked 1 st on rainfall a-coefficients (.2768) and this is followed by Shinyanga area by.624 a-values. The least area is Dodoma with the lowest negative (-.6) a-value. This suggests that there is a high rainfall deficit trend in the area. Mean Maximum Temperatures The results would be observed that Rufiji and Tanga areas are strongly influenced by maximum temperatures. High maximum temperatures usually affect grazing animals on the Coastal area. Some attempts have been made to breed and select lines that could perform better on the Coastal zone (Kavana and Msangi, 25). The least area influenced by maximum temperatures is Dodoma. No apparent reason that could be advanced for the observed conditions. However, Dodoma area is one of the areas where cattle thrive bestthe problem is feed, but the environment is relatively free from livestock diseases. Mean Minimum Temperatures The results would be observed that Shinyanga area is strongly influenced by minimum temperatures. High minimum temperatures usually affect slow the rate of growth in high plants and therefore affect cattle population indirectly through feed availability (Whiteman, 1997). The least area influenced by minimum temperatures is Dodoma. No apparent reason that could be advanced for the observed conditions. However, Dodoma area is one of the areas where cattle thrive best-the problem is feed, but the environment is relatively free from livestock diseases. The correlation strength is in agreement with Brent Sohngen ( who reported that nighttime temperatures over land have increased more than daytime temperatures. Shinyanga ranks top followed by Dodoma, Tanga, Mbeya, Kilimanjaro and Rufiji. All the value for the areas is positive. The b-coefficients are all positive except for Kilimanjaro Area. This suggests that there is a negative trend in the change of cattle population in the area. 23